Skip to main content

Dell Precision T5600: Two Eight-Core CPUs In A Workstation

Results: Adobe Creative Cloud

After Effects

We have a long-standing issue with our After Effects test, where it doesn't respond positively to more than four processing cores. In fact, we've seen instances where adding more compute horsepower actually yields a negative effect.

As you can see, with nine cores (the maximum we can allocate on the Precision T5600 workstation due to its 16 GB of memory) the workload takes two seconds longer to complete than with four cores. Hours have gone into troubleshooting this since Chris first noticed it, and I finally figured out the source of the issue: QuickTime. The source for the three window overlays are QuickTime footage encoded using the QuickTime PNG codec, making them effectively lossless. So, I changed them out for PNG image sequences.

The result is a slight improvement on the baseline machine, but a huge improvement favoring Dell's Precision T5600. It literally renders the test in one-third of the time using the same nine cores.

The side effect, of course, is that the After Effects motion graphics test we've been using to benchmark is finishing too fast on the latest hardware. I decided to increase the workload's demands by upping the project to 1080p. This was achieved by using Illustrator to convert the background graphic into a structured drawing, and then resizing and repositioning the video overlays for a 16:9 aspect ratio.

This was a good idea at first. And it worked great until I moved the test from the baseline machine's hard drive to an SSD, at which point the complete time went from over two minutes to 71 seconds. Then I ran it on the Dell workstation, which chewed through the task in 24 seconds. That's just one extra second more than the 720p-based benchmark, even though the frame being rendered is six times larger.

On the other hand, the HD version is now much more sensitive to storage performance. That's not a bad thing, given how hard it can be to demonstrate the real-world performance benefits of SSDs. Given more system memory, the T5600 could have utilized its additional execution cores, and the test might have finished even faster.

Adobe Premiere

This test involves our standard Hollywood Sequence project featuring DVCProHD 720p footage. Since the footage is only 100 Mb/s, it isn't as sensitive to storage performance. And as a result of optimizations for GPU acceleration, it's much more affected by graphics performance. Remember, this is the latest version of Premiere Pro, so it is OpenCL-enabled, so it also works with AMD's cards.

The encode to H.264, however, is still an entirely CPU-based affair, which is more taxing. Previously, I was rendering transitions and encoding separately so that both tasks could be evaluated on their own. But to make my tests consistent with Premiere benchmarks on other hardware, I used the same settings. Still, the Dell proves itself just under three times faster than our baseline.

Adobe Photoshop

Our two Photoshop tests are based on a series of filters applied to a very large image. They were selected for their prodigious use of threading (in the case of the CPU test) and OpenCL acceleration (for that corresponding benchmark).

The CPU-based Photoshop test falls right about where we'd expect it, just under three times faster over the baseline machine. But the OpenCL result is more difficult to explain. This behavior shows up in our WinZip tests, too. I tried a couple of different Quadro driver revisions and saw little or no difference.

  • kennai
    Would it be possible for you guys to test this in gaming applications? I was really curious how well these CPU's would do in gaming with high end gaming GPU's, since it's pretty much my dream CPU set up >.>.

    Also, good job on the review as always.
    Reply
  • tuffjuff
    So here's what I don't get. With ALL that CPU power, why only 16GB of RAM?
    Reply
  • Am I reading this right, in the SPECviewperf 11 bench graph: the ($480-ish) PNY Quadro 2000 (P500X) beat the ($ 1800-ish) PNY Quadro K5000 by significant margins in the SW-02, as well as some other ones as well. This sure has makes me think twice about wanting to upgrade my 2000 to a K4000.
    Reply
  • blackjackedy
    11768418 said:
    Am I reading this right, in the SPECviewperf 11 bench graph: the ($480-ish) PNY Quadro 2000 (P500X) beat the ($ 1800-ish) PNY Quadro K5000 by significant margins in the SW-02, as well as some other ones as well. This sure has makes me think twice about wanting to upgrade my 2000 to a K4000.

    It says this right beneath the graph:

    The tests seem evenly split between single- and multi-threaded workloads, and some of them incur little or no hit from AA, which points to something other than the GPU bottlenecking performance. In fact, SolidWorks performs better with AA on. How odd is that?

    Reply
  • antemon
    You know, why isn't this sexy casing available in non-business models?
    Reply
  • 11768444 said:
    11768418 said:
    Am I reading this right, in the SPECviewperf 11 bench graph: the ($480-ish) PNY Quadro 2000 (P500X) beat the ($ 1800-ish) PNY Quadro K5000 by significant margins in the SW-02, as well as some other ones as well. This sure has makes me think twice about wanting to upgrade my 2000 to a K4000.

    It says this right beneath the graph:

    The tests seem evenly split between single- and multi-threaded workloads, and some of them incur little or no hit from AA, which points to something other than the GPU bottlenecking performance. In fact, SolidWorks performs better with AA on. How odd is that?
    Correct if I am wrong, but as far as I know the basic S*#tWorks is not optimized for multi-threading (hence I am only running an i7 3820 and anything higher would not benefit the performance). Now SW Simulations and PhotoView360 is a different story.

    I just might run SpecviewPerf 11 on my system to see how it performs. To others it might matter, but in my design, I could care less about AA; I am just happy when SolidWorks does not crash.
    Reply
  • Draven35
    Yes, several of the tests the P500X's higher CPU speed makes a huge difference. Also, ViewPerf uses Solidworks 2010 code, AFAIK.

    Photoview 360's renderer is written by the guys at Luxology, based on the renderer from their 3d application Modo, and is very well multithreaded.

    Tuffjuff: I asked myself the same question about the RAM. The machine would have performed vastly better in the AE tests with 32 GB, because i could have used all of the physical CPU cores.
    Reply
  • bambiboom
    Gentlemen?,
    A very good and welcome review. The systems compared were, however, not at the same level relative to their categories. More would have been revealed if the P500X used something like a GTX 680 (In other words,about 2nd from the top of their respective lines) rather than a Quadro 2000 which is two generations past and in effect, just a much lower line ancestor of the K5000. I imagine these tests are complex and time-consuming, but it would have provided perspective if at least one direct competitor from HP and/or Lenovo appeared.

    A couple of comments on the T5600 design.

    1. I can understand the trends toward more compact cases, and even the need to pander to styling and branding, but the TX600 series is inexecusably short on drive bays. My mother's 2010 dual-core Athlon X2 in a $39 case, "Grandma's TurboKitten 3000", has more expansion bays. Still, the T5600 situation is better than the impending Mac Dustbin Pro.

    2. The brutalist architecture may have convenient handles, but to me is a clunker, both visually and in features. I don't know anyone in architecture, industrial design, graphic design, animation, or video editing that doesn't keep their workstation vertically, who doesn't also hate vertical optical drives, and also often have two of those plus a card reader. Also, As Jon Carroll mentions, this is short on front USB 3.0 ports. I would question a workstation at this level without at least three USB 3.0 ports on the front. There are never enough USB ports on a workstation. The Precision T5400 has two front, six rear, and two on the back of the (SK-8135) keyboard! USB 2.0 ports and I still have to add a four-port hub on one of the back ports.

    Oh, and Jon, the indentation on the top of the T5600 is not for car keys- that's where you would set your short-cabled USB external drive(s)- and flash drives-if there were enough USB 3.0 ports. My Precision T5400 I think is wearing in an indentation in that exact location from a WD Passport.

    3. As tuffjuff also comments, 16GB of RAM is not nearly enough for this kind of machine. Dual CPU systems divide the RAM equally between the processors- these motherboards have separate slots and special sequences of symmetrical positioning. This means that the test system had, in effect, only 8GB of RAM per CPU or as I like to express it- 1GB per core. There's a reason the T5600 .supports 128GB and the T7600 can use 512GB of RAM- Windows, programs and files are big and in these systems, a lot of programs are running at once. I use a formula of 3GB for the OS, 2GB for each simultaneous application and 3GB for open files. As my workstations often use five or six applications plus a constant Intertubes and Windows Exploder, sorry, Explorer, my new four-core HP z420 has 24GB of RAM (6GB/core). If I had a dual E5-2687w system, given there are so many more cores to feed, I would therefore consider 64GB a reasonable level- 32GB per CPU (4GB/core).

    4. The most worrying comments in the review concerns the noise. Of course, a system with two 150W CPU's and school bus- sized GPU needs good airflow, but this one devotes so much of the facade to the grille that the optical drive has to be in the stupid vertical position, and apparently this openness that lets the air in also lets the noise out. But, in my view, noise from a workstation is close to being a deal-breaker. This is another reason why the vertical drive is so silly- few put their workstation horizontally on the desktop right in front of them because of the noise.

    Dell apparently wants to ease out of the declining PC business, and these kinds of design decisions might help that process. I think though that Dell, plus Autodesk and Adobe that want to force eternal cloud computing subscription fees are going to find many, many workstation users that will object and going to buy AutoCad 2014 and CS6, run them on Precision T7500's, and preserve the DVD's in hermetically sealed containers. I, for one, will never, ever be sending my industrial design files into the ether and onto other firms' servers.

    This assessment is a good demonstration of the way in which workstations and creation applications continue to evolve each other. However, as many workstations applications have become far more capable, especially in 3D modeling and simulation, there is still a vast under-utilization of multiple cores in those applications. It's not accidental that the T5600 review emphasized rendering as that it's an example where the core applications have adapted to the availability of multiple cores and also can take advantage of GPU co-processing. It's an odd thing and a puzzle> make a model in Maya and run simulations in Solidworks or Inventor essentially on a single core, but make a rendering of that model using fourteen cores. I make Sketchup Pro models that when they go over about 20MB become almost unusable without navigating in monochrome and clever, careful, and constant fussing with layers. Rendering is very calculation intensive, but so are thermal, gas flow, atmospheric, molecular biological, and structural modeling and simulations.

    The T5600 review, as it's concentrates on applications that reveal the whole capabilities of the $4,000 of CPU's and $1,800 of CUDA cores also reveals this fundamental engineering hollow in workstation applications > and indeed in another important realm. I'm not a gamer, but on this site I can feel gamers wondering the same thing as workstation wonks > Software companies > there are billions of CPU cores waiting for something to do! Why the hell aren't there more multi-core applications?
    Cheers,

    BambiBoom

    PS>

    1. Dell Precision T5400 (2009)> 2X Xeon X5460 quad core @3.16GHz > 16 GB ECC 667> Quadro FX 4800 (1.5GB) > WD RE4 / Segt Brcda 500GB > Windows 7 Ultimate 64-bit > HP 2711x 27" 1920 x 1080 > AutoCad, Revit, Solidworks, Sketchup Pro, Corel Technical Designer, Adobe CS MC, WordP Office, MS Office > architecture, industrial design, graphic design, rendering, writing

    2. HP z420 (2013)> Xeon E5-1620 quad core @ 3.6 / 3.8GHz > 24GB ECC 1600 > Firepro V4900 (Soon Quadro K4000) > Samsung 840 SSD 250GB / Seagate Barracuda 500GB > Windows 7 Professional 64 > to be loaded > AutoCad, Revit, Inventor, Maya (2011), Solidworks 2010, Adobe CS4, Corel Technical Design X-5, Sketchup Pro, WordP Office X-5, MS Office






    Reply
  • Shankovich
    My school updated our lab with these. We run CFD or FEA on them mostly, and it's godly.
    Reply
  • chrpai
    It sure would have been nice to see Visual Studio compile times of Google Chrome.
    Reply